LoDoPaB-CT, a benchmark dataset for low-dose computed tomography reconstruction

Deep learning-based segmentation of ultra-low-dose CT images using an optimized nnU-Net model

PURPOSE

Low-dose CT protocols are widely used for emergency imaging, follow-ups, and attenuation correction in hybrid PET/CT and SPECT/CT imaging. However, low-dose CT images often suffer from reduced quality depending on acquisition and patient attenuation parameters. Deep learning (DL)-based organ segmentation models are typically trained on high-quality images, with limited dedicated models for noisy CT images. This study aimed to develop a DL pipeline for organ segmentation on ultra-low-dose CT images.

MATERIALS AND METHODS

274 CT raw datasets were reconstructed using Siemens ReconCT software with ADMIRE iterative algorithm, generating full-dose (FD-CT) and simulated low-dose (LD-CT) images at 1%, 2%, 5%, and 10% of the original tube current. Existing FD-nnU-Net models segmented 22 organs on FD-CT images, serving as reference masks for training new LD-nnU-Net models using LD-CT images. Three models were trained for bony tissue (6 organs), soft-tissue (15 organs), and body contour segmentation. The segmented masks from LD-CT were compared to FD-CT as standard of reference. External datasets with actual LD-CT images were also segmented and compared.

RESULTS

FD-nnU-Net performance declined with reduced radiation dose, especially below 10% (5 mAs). LD-nnU-Net achieved average Dice scores of 0.937 ± 0.049 (bony tissues), 0.905 ± 0.117 (soft-tissues), and 0.984 ± 0.023 (body contour). LD models outperformed FD models on external datasets.

CONCLUSION

Conventional FD-nnU-Net models performed poorly on LD-CT images. Dedicated LD-nnU-Net models demonstrated superior performance across cross-validation and external evaluations, enabling accurate segmentation of ultra-low-dose CT images. The trained models are available on our GitHub page.

Deep learning-driven pulmonary artery and vein segmentation reveals demography-associated vasculature anatomical differences

Pulmonary artery-vein segmentation is critical for disease diagnosis and surgical planning. Traditional methods rely on Computed Tomography Pulmonary Angiography (CTPA), which requires contrast agents with potential health risks. Non-contrast CT, a safer and more widely available approach, however, has long been considered impossible for this task. Here we propose High-abundant Pulmonary Artery-vein Segmentation (HiPaS), enabling accurate segmentation across both non-contrast CT and CTPA at multiple resolutions. HiPaS integrates spatial normalization with an iterative segmentation strategy, leveraging lower-level vessel segmentations as priors for higher-level segmentations. Trained on a multi-center dataset comprising 1073 CT volumes with manual annotations, HiPaS achieves superior performance (dice score: 91.8%, sensitivity: 98.0%) and demonstrates non-inferiority on non-contrast CT compared to CTPA. Furthermore, HiPaS enables large-scale analysis of 11,784 participants, revealing associations between vessel abundance and sex, age, and diseases, under lung-volume control. HiPaS represents a promising, non-invasive approach for clinical diagnostics and anatomical research.

A nonlocal prior in iterative CT reconstruction

BACKGROUND

Computed tomography (CT) reconstruction problems are always framed as inverse problems, where the attenuation map of an imaged object is reconstructed from the sinogram measurement. In practice, these inverse problems are often ill-posed, especially under few-view and limited-angle conditions, which makes accurate reconstruction challenging. Existing solutions use regularizations such as total variation to steer reconstruction algorithms to the most plausible result. However, most prevalent regularizations rely on the same priors, such as piecewise constant prior, hindering their ability to collaborate effectively and further boost reconstruction precision.

PURPOSE

This study aims to overcome the aforementioned challenge a prior previously limited to discrete tomography. This enables more accurate reconstructions when the proposed method is used in conjunction with most existing regularizations as they utilize different priors. The improvements will be demonstrated through experiments conducted under various conditions.

METHODS

Inspired by the discrete algebraic reconstruction technique (DART) algorithm for discrete tomography, we find out that pixel grayscale values in CT images are not uniformly distributed and are actually highly clustered. Such discovery can be utilized as a powerful prior for CT reconstruction. In this paper, we leverage the collaborative filtering technique to enable the collaboration of the proposed prior and most existing regularizations, significantly enhancing the reconstruction accuracy.

RESULTS

Our experiments show that the proposed method can work with most existing regularizations and significantly improve the reconstruction quality. Such improvement is most pronounced under limited-angle and few-view conditions. Furthermore, the proposed regularization also has the potential for further improvement and can be utilized in other image reconstruction areas.

CONCLUSIONS

We propose improving the performance of iterative CT reconstruction algorithms by applying the collaborative filtering technique along with a prior based on the densely clustered distribution of pixel grayscale values in CT images. Our experimental results indicate that the proposed methodology consistently enhances reconstruction accuracy when used in conjunction with most existing regularizations, particularly under few-view and limited-angle conditions.

Conditioning generative latent optimization for sparse-view computed tomography image reconstruction

Purpose

The issue of delivered doses during computed tomography (CT) scans encouraged sparser sets of X-ray projection, severely degrading reconstructions from conventional methods. Although most deep learning approaches benefit from large supervised datasets, they cannot generalize to new acquisition protocols (geometry, source/detector specifications). To address this issue, we developed a method working without training data and independently of experimental setups. In addition, our model may be initialized on small unsupervised datasets to enhance reconstructions.

Approach

We propose a conditioned generative latent optimization (cGLO) in which a decoder reconstructs multiple slices simultaneously with a shared objective. It is tested on full-dose sparse-view CT for varying projection sets: (a) without training data against Deep Image Prior (DIP) and (b) with training datasets of multiple sizes against state-of-the-art score-based generative models (SGMs). Peak signal-to-noise ratio (PSNR) and structural SIMilarity (SSIM) metrics are used to quantify reconstruction quality.

Results

cGLO demonstrates better SSIM than SGMs (between + 0.034 and + 0.139 ) and has an increasing advantage for smaller datasets reaching a + 6.06    dB PSNR gain. Our strategy also outperforms DIP with at least a + 1.52    dB PSNR advantage and peaks at + 3.15    dB with fewer angles. Moreover, cGLO does not create artifacts or structural deformations contrary to DIP and SGMs.

Conclusions

We propose a parsimonious and robust reconstruction technique offering similar to better performances when compared with state-of-the-art methods regarding full-dose sparse-view CT. Our strategy could be readily applied to any imaging reconstruction task without any assumption about the acquisition protocol or the quantity of available data.

Artificial Intelligence in Computed Tomography Image Reconstruction: A Review of Recent Advances

The development of novel image reconstruction algorithms has been pivotal in enhancing image quality and reducing radiation dose in computed tomography (CT) imaging. Traditional techniques like filtered back projection perform well under ideal conditions but fail to generate high-quality images under low-dose, sparse-view, and limited-angle conditions. Iterative reconstruction methods improve upon filtered back projection by incorporating system models and assumptions about the patient, yet they can suffer from patchy image textures. The emergence of artificial intelligence (AI), particularly deep learning, has further advanced CT reconstruction. AI techniques have demonstrated great potential in reducing radiation dose while preserving image quality and noise texture. Moreover, AI has exhibited unprecedented performance in addressing challenging CT reconstruction problems, including low-dose CT, sparse-view CT, limited-angle CT, and interior tomography. This review focuses on the latest advances in AI-based CT reconstruction under these challenging conditions.

Structure-aware diffusion for low-dose CT imaging

Reducing the radiation dose leads to the x-ray computed tomography (CT) images suffering from heavy noise and artifacts, which inevitably interferes with the subsequent clinic diagnostic and analysis. Leading works have explored diffusion models for low-dose CT imaging to avoid the structure degeneration and blurring effects of previous deep denoising models. However, most of them always begin their generative processes with Gaussian noise, which has little or no structure priors of the clean data distribution, thereby leading to long-time inference and unpleasant reconstruction quality. To alleviate these problems, this paper presents a Structure-Aware Diffusion model (SAD), an end-to-end self-guided learning framework for high-fidelity CT image reconstruction. First, SAD builds a nonlinear diffusion bridge between clean and degraded data distributions, which could directly learn the implicit physical degradation prior from observed measurements. Second, SAD integrates the prompt learning mechanism and implicit neural representation into the diffusion process, where rich and diverse structure representations extracted by degraded inputs are exploited as prompts, which provides global and local structure priors, to guide CT image reconstruction. Finally, we devise an efficient self-guided diffusion architecture using an iterative updated strategy, which further refines structural prompts during each generative step to drive finer image reconstruction. Extensive experiments on AAPM-Mayo and LoDoPaB-CT datasets demonstrate that our SAD could achieve superior performance in terms of noise removal, structure preservation, and blind-dose generalization, with few generative steps, even one step only.

Low-dose CT reconstruction using dataset-free learning

Low-Dose computer tomography (LDCT) is an ideal alternative to reduce radiation risk in clinical applications. Although supervised-deep-learning-based reconstruction methods have demonstrated superior performance compared to conventional model-driven reconstruction algorithms, they require collecting massive pairs of low-dose and norm-dose CT images for neural network training, which limits their practical application in LDCT imaging. In this paper, we propose an unsupervised and training data-free learning reconstruction method for LDCT imaging that avoids the requirement for training data. The proposed method is a post-processing technique that aims to enhance the initial low-quality reconstruction results, and it reconstructs the high-quality images by neural work training that minimizes the ℓ1-norm distance between the CT measurements and their corresponding simulated sinogram data, as well as the total variation (TV) value of the reconstructed image. Moreover, the proposed method does not require to set the weights for both the data fidelity term and the plenty term. Experimental results on the AAPM challenge data and LoDoPab-CT data demonstrate that the proposed method is able to effectively suppress the noise and preserve the tiny structures. Also, these results demonstrate the rapid convergence and low computational cost of the proposed method. The source code is available at https://github.com/linfengyu77/IRLDCT.

DDT-Net: Dose-Agnostic Dual-Task Transfer Network for Simultaneous Low-Dose CT Denoising and Simulation

Deep learning (DL) algorithms have achieved unprecedented success in low-dose CT (LDCT) imaging and are expected to be a new generation of CT reconstruction technology. However, most DL-based denoising models often lack the ability to generalize to unseen dose data. Moreover, most simulation tools for LDCT typically operate on proprietary projection data, which is generally not accessible without an established collaboration with CT manufacturers. To alleviate these issues, in this work, we propose a dose-agnostic dual-task transfer network, termed DDT-Net, for simultaneous LDCT denoising and simulation. Concretely, the dual-task learning module is constructed to integrate the LDCT denoising and simulation tasks into a unified optimization framework by learning the joint distribution of LDCT and NDCT data. We approximate the joint distribution of continuous dose level data by training DDT-Net with discrete dose data, which can be generalized to denoising and simulation of unseen dose data. In particular, the mixed-dose training strategy adopted by DDT-Net can promote the denoising performance of lower-dose data. The paired dataset simulated by DDT-Net can be used for data augmentation to further restore the tissue texture of LDCT images. Experimental results on synthetic data and clinical data show that the proposed DDT-Net outperforms competing methods in terms of denoising and generalization performance at unseen dose data, and it also provides a simulation tool that can quickly simulate realistic LDCT images at arbitrary dose levels.

Noise-Generating and Imaging Mechanism Inspired Implicit Regularization Learning Network for Low Dose CT Reconstrution

Low-dose computed tomography (LDCT) helps to reduce radiation risks in CT scanning while maintaining image quality, which involves a consistent pursuit of lower incident rays and higher reconstruction performance. Although deep learning approaches have achieved encouraging success in LDCT reconstruction, most of them treat the task as a general inverse problem in either the image domain or the dual (sinogram and image) domains. Such frameworks have not considered the original noise generation of the projection data and suffer from limited performance improvement for the LDCT task. In this paper, we propose a novel reconstruction model based on noise-generating and imaging mechanism in full-domain, which fully considers the statistical properties of intrinsic noises in LDCT and prior information in sinogram and image domains. To solve the model, we propose an optimization algorithm based on the proximal gradient technique. Specifically, we derive the approximate solutions of the integer programming problem on the projection data theoretically. Instead of hand-crafting the sinogram and image regularizers, we propose to unroll the optimization algorithm to be a deep network. The network implicitly learns the proximal operators of sinogram and image regularizers with two deep neural networks, providing a more interpretable and effective reconstruction procedure. Numerical results demonstrate our proposed method improvements of > 2.9 dB in peak signal to noise ratio, > 1.4% promotion in structural similarity metric, and > 9 HU decrements in root mean square error over current state-of-the-art LDCT methods.

Hierarchical decomposed dual-domain deep learning for sparse-view CT reconstruction

Objective. X-ray computed tomography employing sparse projection views has emerged as a contemporary technique to mitigate radiation dose. However, due to the inadequate number of projection views, an analytic reconstruction method utilizing filtered backprojection results in severe streaking artifacts. Recently, deep learning (DL) strategies employing image-domain networks have demonstrated remarkable performance in eliminating the streaking artifact caused by analytic reconstruction methods with sparse projection views. Nevertheless, it is difficult to clarify the theoretical justification for applying DL to sparse view computed tomography (CT) reconstruction, and it has been understood as restoration by removing image artifacts, not reconstruction.Approach. By leveraging the theory of deep convolutional framelets (DCF) and the hierarchical decomposition of measurement, this research reveals the constraints of conventional image and projection-domain DL methodologies, subsequently, the research proposes a novel dual-domain DL framework utilizing hierarchical decomposed measurements. Specifically, the research elucidates how the performance of the projection-domain network can be enhanced through a low-rank property of DCF and a bowtie support of hierarchical decomposed measurement in the Fourier domain.Main results. This study demonstrated performance improvement of the proposed framework based on the low-rank property, resulting in superior reconstruction performance compared to conventional analytic and DL methods.Significance. By providing a theoretically justified DL approach for sparse-view CT reconstruction, this study not only offers a superior alternative to existing methods but also opens new avenues for research in medical imaging. It highlights the potential of dual-domain DL frameworks to achieve high-quality reconstructions with lower radiation doses, thereby advancing the field towards safer and more efficient diagnostic techniques. The code is available athttps://github.com/hanyoseob/HDD-DL-for-SVCT.

Self-supervised dual-domain balanced dropblock-network for low-dose CT denoising

Objective.Self-supervised learning methods have been successfully applied for low-dose computed tomography (LDCT) denoising, with the advantage of not requiring labeled data. Conventional self-supervised methods operate only in the image domain, ignoring valuable priors in the sinogram domain. Recently proposed dual-domain methods address this limitation but encounter issues with blurring artifacts in the reconstructed image due to the inhomogeneous distribution of noise levels in low-dose sinograms.Approach.To tackle this challenge, this paper proposes SDBDNet, an end-to-end dual-domain self-supervised method for LDCT denoising. With the network designed based on the properties of inhomogeneous noise in low-dose sinograms and the principle of moderate sinogram-domain denoising, SDBDNet achieves effective denoising in dual domains without introducing blurring artifacts. Specifically, we split the sinogram into two subsets based on the positions of detector cells to generate paired training data with high similarity and independent noise. These sub-sinograms are then restored to their original size using 1D interpolation and learning-based correction. To achieve adaptive and moderate smoothing in the sinogram domain, we integrate Dropblock, a type of convolution layer with regularization, into SDBDNet, and set a weighted average between the denoised sinograms and their noisy counterparts, leading to a well-balanced dual-domain approach.Main results.Numerical experiments show that our method outperforms popular non-learning and self-supervised learning methods, demonstrating its effectiveness and superior performance.Significance.While introducing a novel high-performance dual-domain self-supervised LDCT denoising method, this paper also emphasizes and verifies the importance of appropriate sinogram-domain denoising in dual-domain methods, which might inspire future work.

APNet: Adaptive projection network for medical image denoising

BACKGROUND

In clinical medicine, low-dose radiographic image noise reduces the quality of the detected image features and may have a negative impact on disease diagnosis.

OBJECTIVE

In this study, Adaptive Projection Network (APNet) is proposed to reduce noise from low-dose medical images.

METHODS

APNet is developed based on an architecture of the U-shaped network to capture multi-scale data and achieve end-to-end image denoising. To adaptively calibrate important features during information transmission, a residual block of the dual attention method throughout the encoding and decoding phases is integrated. A non-local attention module to separate the noise and texture of the image details by using image adaptive projection during the feature fusion.

RESULTS

To verify the effectiveness of APNet, experiments on lung CT images with synthetic noise are performed, and the results demonstrate that the proposed approach outperforms recent methods in both quantitative index and visual quality. In addition, the denoising experiment on the dental CT image is also carried out and it verifies that the network has a certain generalization.

CONCLUSIONS

The proposed APNet is an effective method that can reduce image noise and preserve the required image details in low-dose radiographic images.

A Deep Convolutional Gated Recurrent Unit for CT Image Reconstruction

Computed tomography (CT) is one of the most important medical imaging technologies in use today. Most commercial CT products use a technique known as the filtered backprojection (FBP) that is fast and can produce decent image quality when an X-ray dose is high. However, the FBP is not good enough on low-dose X-ray CT imaging because the CT image reconstruction problem becomes more stochastic. A more effective reconstruction technique proposed recently and implemented in a limited number of CT commercial products is an iterative reconstruction (IR). The IR technique is based on a Bayesian formulation of the CT image reconstruction problem with an explicit model of the CT scanning, including its stochastic nature, and a prior model that incorporates our knowledge about what a good CT image should look like. However, constructing such prior knowledge is more complicated than it seems. In this article, we propose a novel neural network for CT image reconstruction. The network is based on the IR formulation and constructed with a recurrent neural network (RNN). Specifically, we transform the gated recurrent unit (GRU) into a neural network performing CT image reconstruction. We call it "GRU reconstruction." This neural network conducts concurrent dual-domain learning. Many deep learning (DL)-based methods in medical imaging are single-domain learning, but dual-domain learning performs better because it learns from both the sinogram and the image domain. In addition, we propose backpropagation through stage (BPTS) as a new RNN backpropagation algorithm. It is similar to the backpropagation through time (BPTT) of an RNN; however, it is tailored for iterative optimization. Results from extensive experiments indicate that our proposed method outperforms conventional model-based methods, single-domain DL methods, and state-of-the-art DL techniques in terms of the root mean squared error (RMSE), the peak signal-to-noise ratio (PSNR), and the structure similarity (SSIM) and in terms of visual appearance.

Unsupervised learning-based dual-domain method for low-dose CT denoising

Objective. Low-dose CT (LDCT) is an important research topic in the field of CT imaging because of its ability to reduce radiation damage in clinical diagnosis. In recent years, deep learning techniques have been widely applied in LDCT imaging and a large number of denoising methods have been proposed. However, One major challenge of supervised deep learning-based methods is the exactly geometric pairing of datasets with different doses. Therefore, the aim of this study is to develop an unsupervised learning-based LDCT imaging method to address the aforementioned challenges.Approach. In this paper, we propose an unsupervised learning-based dual-domain method for LDCT denoising, which consists of two stages: the first stage is projection domain denoising, in which the unsupervised learning method Noise2Self is applied to denoise the projection data with statistically independent and zero-mean noise. The second stage is an iterative enhancement approach, which combines the prior information obtained from the generative model with an iterative reconstruction algorithm to enhance the details of the reconstructed image.Main results. Experimental results show that our proposed method outperforms the comparison method in terms of denoising effect. Particularly, in terms of SSIM, the denoised results obtained using our method achieve the highest SSIM.Significance. In conclusion, our unsupervised learning-based method can be a promising alternative to the traditional supervised methods for LDCT imaging, especially when the availability of the labeled datasets is limited.

2DeteCT - A large 2D expandable, trainable, experimental Computed Tomography dataset for machine learning

Recent research in computational imaging largely focuses on developing machine learning (ML) techniques for image reconstruction, which requires large-scale training datasets consisting of measurement data and ground-truth images. However, suitable experimental datasets for X-ray Computed Tomography (CT) are scarce, and methods are often developed and evaluated only on simulated data. We fill this gap by providing the community with a versatile, open 2D fan-beam CT dataset suitable for developing ML techniques for a range of image reconstruction tasks. To acquire it, we designed a sophisticated, semi-automatic scan procedure that utilizes a highly-flexible laboratory X-ray CT setup. A diverse mix of samples with high natural variability in shape and density was scanned slice-by-slice (5,000 slices in total) with high angular and spatial resolution and three different beam characteristics: A high-fidelity, a low-dose and a beam-hardening-inflicted mode. In addition, 750 out-of-distribution slices were scanned with sample and beam variations to accommodate robustness and segmentation tasks. We provide raw projection data, reference reconstructions and segmentations based on an open-source data processing pipeline.

CDDnet: Cross-domain denoising network for low-dose CT image via local and global information alignment

The domain shift problem has emerged as a challenge in cross-domain low-dose CT (LDCT) image denoising task, where the acquisition of a sufficient number of medical images from multiple sources may be constrained by privacy concerns. In this study, we propose a novel cross-domain denoising network (CDDnet) that incorporates both local and global information of CT images. To address the local component, a local information alignment module has been proposed to regularize the similarity between extracted target and source features from selected patches. To align the general information of the semantic structure from a global perspective, an autoencoder is adopted to learn the latent correlation between the source label and the estimated target label generated by the pre-trained denoiser. Experimental results demonstrate that our proposed CDDnet effectively alleviates the domain shift problem, outperforming other deep learning-based and domain adaptation-based methods under cross-domain scenarios.

Deep learning based de-overlapping correction of projections from a flat-panel micro array X-ray source: Simulation study

PURPOSE

Flat-panel X-ray source is an experimental X-ray emitter with target application of static computer tomography (CT), which can save imaging space and time. However, the X-ray cone beams emitted by the densely arranged micro-ray sources are overlapped, causing serious structural overlapping and visual blur in the projection results. Traditional deoverlapping methods can hardly solve this problem well.

METHOD

We converted the overlapping cone beam projections to parallel beam projections through a U-like neural network and selected structural similarity (SSIM) loss as the loss function. In this study, we converted three kinds of overlapping cone beam projections of the Shepp-Logan, line-pairs, and abdominal data with two overlapping levels to corresponding parallel beam projections. Training completed, we tested the model using the test set data that was not used at the training phase, and evaluated the difference between the test set conversion results and their corresponding parallel beams through three indicators: mean squared error (MSE), peak signal-to-noise ratio (PSNR) and SSIM. In addition, projections from head phantoms were applied for generalization test.

RESULT

In the Shepp-Logan low-overlapping task, we obtained a MSE of 1.624×10^-5, a PSNR of 47.892 dB, and a SSIM of 0.998 which are the best results of the six experiments. For the most challenging abdominal task, the MSE, PSNR, and SSIM are 1.563×10^-3, 28.0586 dB, and 0.983, respectively. In more generalized data, the model also achieved good results.

CONCLUSION

This study proves the feasibility of utilizing the end-to-end U-net for deblurring and deoverlapping in the flat-panel X-ray source domain.

Transforming medical imaging with Transformers? A comparative review of key properties, current progresses, and future perspectives

Transformer, one of the latest technological advances of deep learning, has gained prevalence in natural language processing or computer vision. Since medical imaging bear some resemblance to computer vision, it is natural to inquire about the status quo of Transformers in medical imaging and ask the question: can the Transformer models transform medical imaging? In this paper, we attempt to make a response to the inquiry. After a brief introduction of the fundamentals of Transformers, especially in comparison with convolutional neural networks (CNNs), and highlighting key defining properties that characterize the Transformers, we offer a comprehensive review of the state-of-the-art Transformer-based approaches for medical imaging and exhibit current research progresses made in the areas of medical image segmentation, recognition, detection, registration, reconstruction, enhancement, etc. In particular, what distinguishes our review lies in its organization based on the Transformer's key defining properties, which are mostly derived from comparing the Transformer and CNN, and its type of architecture, which specifies the manner in which the Transformer and CNN are combined, all helping the readers to best understand the rationale behind the reviewed approaches. We conclude with discussions of future perspectives.

Deep image and feature prior algorithm based on U-ConformerNet structure

PURPOSE

The reconstruction performance of the deep image prior (DIP) approach is limited by the conventional convolutional layer structure and it is difficult to enhance its potential. In order to improve the quality of image reconstruction and suppress artifacts, we propose a DIP algorithm with better performance, and verify its superiority in the latest case.

METHODS

We construct a new U-ConformerNet structure as the DIP algorithm's network, replacing the traditional convolutional layer-based U-net structure, and introduce the 'lpips' deep network based feature distance regularization method. Our algorithm can switch between supervised and unsupervised modes at will to meet different needs.

RESULTS

The reconstruction was performed on the low dose CT dataset (LoDoPaB). Our algorithm attained a PSNR of more than 35 dB under unsupervised conditions, and the PSNR under the supervised condition is greater than 36 dB. Both of which are better than the performance of the DIP-TV. Furthermore, the accuracy of this method is positively connected with the quality of the a priori image with the help of deep networks. In terms of noise eradication and artifact suppression, the DIP algorithm with U-ConformerNet structure outperforms the standard DIP method based on convolutional structure.

CONCLUSIONS

It is known by experimental verification that, in unsupervised mode, the algorithm improves the output PSNR by at least 2-3 dB when compared to the DIP-TV algorithm (proposed in 2020). In supervised mode, our algorithm approaches that of the state-of-the-art end-to-end deep learning algorithms.

An AI-Based Colonic Polyp Classifier for Colorectal Cancer Screening Using Low-Dose Abdominal CT

Among the non-invasive Colorectal cancer (CRC) screening approaches, Computed Tomography Colonography (CTC) and Virtual Colonoscopy (VC), are much more accurate. This work proposes an AI-based polyp detection framework for virtual colonoscopy (VC). Two main steps are addressed in this work: automatic segmentation to isolate the colon region from its background, and automatic polyp detection. Moreover, we evaluate the performance of the proposed framework on low-dose Computed Tomography (CT) scans. We build on our visualization approach, Fly-In (FI), which provides "filet"-like projections of the internal surface of the colon. The performance of the Fly-In approach confirms its ability with helping gastroenterologists, and it holds a great promise for combating CRC. In this work, these 2D projections of FI are fused with the 3D colon representation to generate new synthetic images. The synthetic images are used to train a RetinaNet model to detect polyps. The trained model has a 94% f1-score and 97% sensitivity. Furthermore, we study the effect of dose variation in CT scans on the performance of the the FI approach in polyp visualization. A simulation platform is developed for CTC visualization using FI, for regular CTC and low-dose CTC. This is accomplished using a novel AI restoration algorithm that enhances the Low-Dose CT images so that a 3D colon can be successfully reconstructed and visualized using the FI approach. Three senior board-certified radiologists evaluated the framework for the peak voltages of 30 KV, and the average relative sensitivities of the platform were 92%, whereas the 60 KV peak voltage produced average relative sensitivities of 99.5%.

Iterative tomographic reconstruction with TV prior for low-dose CBCT dental imaging

Objective.Cone-beam computed tomography is becoming more and more popular in applications such as 3D dental imaging. Iterative methods compared to the standard Feldkamp algorithm have shown improvements in image quality of reconstruction of low-dose acquired data despite their long computing time. An interesting aspect of iterative methods is their ability to include prior information such as sparsity-constraint. While a large panel of optimization algorithms along with their adaptation to tomographic problems are available, they are mainly studied on 2D parallel or fan-beam data. The issues raised by 3D CBCT and moreover by truncated projections are still poorly understood.Approach.We compare different carefully designed optimization schemes in the context of realistic 3D dental imaging. Besides some known algorithms, SIRT-TV and MLEM, we investigate the primal-dual hybrid gradient (PDHG) approach and a newly proposed MLEM-TV optimizer. The last one is alternating EM steps and TV-denoising, combination not yet investigated for CBCT. Experiments are performed on both simulated data from a 3D jaw phantom and data acquired with a dental clinical scanner.Main results.With some adaptations to the specificities of CBCT operators, PDHG and MLEM-TV algorithms provide the best reconstruction quality. These results were obtained by comparing the full-dose image with a low-dose image and an ultra low-dose image.Significance.The convergence speed of the original iterative methods is hampered by the conical geometry and significantly reduced compared to parallel geometries. We promote the pre-conditioned version of PDHG and we propose a pre-conditioned version of the MLEM-TV algorithm. To the best of our knowledge, this is the first time PDHG and convergent MLEM-TV algorithms are evaluated on experimental dental CBCT data, where constraints such as projection truncation and presence of metal have to be jointly overcome.

Unsupervised knowledge-transfer for learned image reconstruction

Deep learning-based image reconstruction approaches have demonstrated impressive empirical performance in many imaging modalities. These approaches usually require a large amount of high-quality paired training data, which is often not available in medical imaging. To circumvent this issue we develop a novel unsupervised knowledge-transfer paradigm for learned reconstruction within a Bayesian framework. The proposed approach learns a reconstruction network in two phases. The first phase trains a reconstruction network with a set of ordered pairs comprising of ground truth images of ellipses and the corresponding simulated measurement data. The second phase fine-tunes the pretrained network to more realistic measurement data without supervision. By construction, the framework is capable of delivering predictive uncertainty information over the reconstructed image. We present extensive experimental results on low-dose and sparse-view computed tomography showing that the approach is competitive with several state-of-the-art supervised and unsupervised reconstruction techniques. Moreover, for test data distributed differently from the training data, the proposed framework can significantly improve reconstruction quality not only visually, but also quantitatively in terms of PSNR and SSIM, when compared with learned methods trained on the synthetic dataset only.

A blind medical image denoising method with noise generation network

BACKGROUND

In the process of medical images acquisition, the unknown mixed noise will affect image quality. However, the existing denoising methods usually focus on the known noise distribution.

OBJECTIVE

In order to remove the unknown real noise in low-dose CT images (LDCT), a two-step deep learning framework is proposed in this study, which is called Noisy Generation-Removal Network (NGRNet).

METHODS

Firstly, the output results of L0 Gradient Minimization are used as the labels of a dental CT image dataset to form a pseudo-image pair with the real dental CT images, which are used to train the noise generation network to estimate real noise distribution. Then, for the lung CT images of the LIDC/IDRI database, we migrate the real noise to the noise-free lung CT images, to construct a new almost-real noisy images dataset. Since dental images and lung images are all CT images, this migration can be achieved. The denoising network is trained to realize the denoising of real LDCT for dental images by using this dataset but can extend for any low-dose CT images.

RESULTS

To prove the effectiveness of our NGRNet, we conduct experiments on lung CT images with synthetic noise and tooth CT images with real noise. For synthetic noise image datasets, experimental results show that NGRNet is superior to existing denoising methods in terms of visual effect and exceeds 0.13dB in the peak signal-to-noise ratio (PSNR). For real noisy image datasets, the proposed method can achieve the best visual denoising effect.

CONCLUSIONS

The proposed method can retain more details and achieve impressive denoising performance.

Conditional Invertible Neural Networks for Medical Imaging

Over recent years, deep learning methods have become an increasingly popular choice for solving tasks from the field of inverse problems. Many of these new data-driven methods have produced impressive results, although most only give point estimates for the reconstruction. However, especially in the analysis of ill-posed inverse problems, the study of uncertainties is essential. In our work, we apply generative flow-based models based on invertible neural networks to two challenging medical imaging tasks, i.e., low-dose computed tomography and accelerated medical resonance imaging. We test different architectures of invertible neural networks and provide extensive ablation studies. In most applications, a standard Gaussian is used as the base distribution for a flow-based model. Our results show that the choice of a radial distribution can improve the quality of reconstructions.